On Sat, 3 May 2025 05:03:36 GMT, Shaojin Wen <s...@openjdk.org> wrote:
>> The byte[] allocated in Integer/Long.toString is fully filled, so we can use
>> StringConcatHelper::newArray to create byte[] to improve performance.
>
> Shaojin Wen has updated the pull request incrementally with one additional
> commit since the last revision:
>
> use Unsafe::allocateUninitializedArray
public static String toString(int i) {
int size = DecimalDigits.stringSize(i);
byte coder = COMPACT_STRINGS ? LATIN1 : UTF16;
byte[] buf = (byte[])
Unsafe.getUnsafe().allocateUninitializedArray(byte.class, size << coder);
if (coder == LATIN1) {
DecimalDigits.getCharsLatin1(i, size, buf);
} else {
DecimalDigits.getCharsUTF16(i, size, buf);
}
return new String(buf, coder);
}
This implementation has a smaller code size and the same performance, but it
changes the code structure, so I did not use it.
-------------
PR Comment: https://git.openjdk.org/jdk/pull/23353#issuecomment-2848749465
